Harter's Self-Perception Profile for Adolescents: reliability, validity, and evaluation of the question format

The Self-Perception Profile for Adolescents (SPPA; Harter, 1988) was administered to a national representative sample of 11,315 Norwegian adolescents of the ages 13 to 20. The original version of SPPA has an idiosyncratic and time-consuming item format, describing two adolescents with opposite characteristics on each item. A revised version of SPPA was developed using only one statement for each item. The original SPPA was administered to a random subsample of 880 subjects, whereas the rest of the total sample completed the revised edition. Concerning the original SPPA, results showed low to adequate reliability and poor replication of the factorial pattern. However, low correlations with the Marlowe-Crowne Social Desirability Index were obtained, and the construct validity was supported. Concerning the revised edition, the a priori factorial pattern was replicated. Further, it had substantially better reliability, better convergent validity, and better factorial validity than the original version. The discriminant validity and the contamination by social desirability bias were the same for both versions. The data thus indicate that there is no need to retain the time-consuming and cumbersome item format of the original SPPA.